OUTPUT_DATA_BAD_TYPE - resulting in loss of rows in data

[nshapir2]

so I have a dataset in an S3 connection that I am enriching with partition information and writing to a HDFS connection. I am losing 1100 records due to bad datatype. 

How can I override this? 

I tried by adding something to the prepare recipe where I use a formula ToNumber(Field1) saved as Field1_. I deleted Field1 then renamed Field1_ to Field1.  That did not change anything

[SarinaS]

Hi @nshapir2,

We would need to review the job diagnostic and possibly a sample of the input data in order to advise on the best steps to resolve the issue, since it seems like there is a problem parsing your underlying data. Please feel free to open a support ticket with a job diagnostic and a sample of the data, and we can review it!

Thanks,
Sarina